HUEY LOUIS KOSTANICK
BALKAN DEMOGRAPHIC TRENDS AND POPULATION HEARTLANDS
The Balkan peninsula juts out from southeastern Europe into the Mediterranean between the Adriatic and Black Seas. This position at the meeting place of Europe, Asia and Africa has given it a transitional role in history, a role associated with easy peripheral accessibility that has brought in a volatile infusion of neighboring peoples, cultures and empires which have left marked centrifugal influences on the peninsula. No less have internal forces tended toward factionalism, imperialism and divism between political entities, creating a legacy of conflicting territorial claims that make the Balkans probably the leading area in the world of 'problem areas', meaning those areas of a state claimed by another state.
These internal territorial conflicts are based to a great degree on the ethnic complexity of the Balkans, complexity born of different races, languages, religions and customs. The erosive clash of the Turkish and Austro-Hungarian empires brought to the fore the necessity of specific identification of ethnic groups for political purposes both within the empires and against the empires. Thus the Balkans became, in a sense, a world laboratory for ethnic group identification, just as territorially it is a classical model of problem areas.
Because the long history of battle for national independence is so absorbing, it becomes a temptation to view the Balkans as an area of static equilibrium — a sort of museum piece of history, often enshrined in period pieces of colorful musical comedy such as Franz Lehar's Merry Widow. Yet, in truth, the same historical processes are in operation now as in the past and are producing changes as strong and as fundamental as those in the past.
TRANSITIONAL TERRITORIAL ROLE — THE CULTURAL IMPRESS
Although the end of World War I brought the emergence of the present Balkan states and the end of the Ottoman, Austro-Hungarian and Russian
425
empires, the cultural imprint of these and earlier occupiers still remains on the Balkan landscape, both physically and culturally. Yugoslavia is a classical example of this as is evident in a transect from northwest to southeast. Slovenia, in the northwest, with the adjacent portions of Croatia and Bosnia-Herzegovina was under Austrian administration. The cultural heritage is still evident in architecture, cuisine, language, railroad patterns and other cultural characteristics. Similarly the Hungarian imprint was left on northeast Yugoslavia, and Turkish influences are clearly seen in southeast Yugoslavia. Both Italian and Roman traces persist in Dalmatia and Istria. In broader regional terms, Albania, Bulgaria and Greece all show marked heritage from the Turkish period and Romania still carries within itself areal vestiges of Hungarian, German and Russian occupation.
Since World War II, the dominance of Communist regimes in Hungary, Romania, Yugoslavia, Albania and Bulgaria has created a new ideological empire which, while technically not analogous to the previous empires, may still in time again create a new historical impress on the landscape that will produce a new contrast between those countries and the non-Communist states of Greece and Turkey. One of the changes on the landscape that is already noticeable is the consolidation of smaller land holdings into large collective farms in Hungary, Romania, and Bulgaria and in certain parts of Yugoslavia and Albania. Urban influences are also discernible in 'socialist' architecture, such as the State Printing House in Bucharest, and in the ubiquitous 'Park of Rest and Culture' and 'Palace of Culture' that dot the cities of Communist-ruled Eastern Europe, as well as the more ephemeral posters and banners that have been so prominent a part of the Communist city.
Sharp distinction must be made between the spread of Communist ideology as opposed to Soviet political domination. Immediately after World War II it appeared that the Soviet Union would gain hegemony of the whole Balkan peninsula through successful Communist penetration of Greece, which would have given the Soviet Union access to the Aegean Sea. [1] At the same time political pressure was placed on Turkey through activity in the Azerbaidjan in Iran. Had both sections of the pincer
1. Hugh Setton-Watson, The East European Revolution (New York and Washington, D. C, Praeger 1956), 318-328. Also Robert Lee Wolff, The Balkans in Our Time (Cambridge: Harvard University Press, 1956), 264-265 and 317. For a vivid account of the Athens battle of 1944-1945 see L. S. Stavrianos, The Balkans since 1453 (New York : Holt. Rinchart and Winston, 1958), 820-830.
426
movement worked, the Soviet Union would have gained probable control of the Turkish Straits and thus easy entrance to the eastern Mediterranean which would have put corresponding pressure on the British strategic control points of Gibraltar and the Suez Canal. This was plainly a power push for Soviet control of the peninsula. But the Azerbaidjan effort was stopped and so was the guerrilla effort in Greece, where aid was first given by the British and then following 1947 by American military and economic aid under the Truman doctrine. Thus in 1947, there seemed to be a clear confrontation between the Soviet Union and the United States in terms of power politics. Had this continued, it would have meant division of the peninsula into two power blocs with Hungary, Romania, Yugoslavia, Bulgaria and Albania in the Soviet sphere of influence and Greece and Turkey in the sphere of influence of the United States.
This would have been a classical pattern of division of the Balkan peninsula into clearly delimited zones dominated by foreign powers, but, in fact, this did not happen because in 1948 there was a political break between Yugoslavia and the Soviet Union, which broke Soviet domination although the supremacy of the Communist structure has still remained in Yugoslavia. [2] This schism was extended by the subsequent shift in the late 1950's of Albania from alliance with the Soviet Union to alliance with Communist China, [3] and in the early 1960's by the assertion of greater independence from the Soviet Union by Romania.
Thus in short order the Balkan Peninsula changed from apparent polarized East-West domination to an area of diverse political entities, which show constantly greater independence and individuality. The extent of change in relationships is exemplified by the signing of a treaty of friendship between Greece, Turkey and Yugoslavia in 1953 [4] which led to a short exchange of state visits with each other until the deterioration of relations between Greece and Turkey over Cyprus.
Hence the Balkan Peninsula still demonstrates its transitional character for not only is it a contact zone of East-West confrontation, it also reflects the conflicts within the Communist group from orthodoxy to 'Titoism' and from alliance with the Soviet Union and the other Communist countries of Eastern Europe to alliance with Communist China.
2. Wolff, op. cit., 352-448.
3. Charles and Barbara Jelavich, The Balkans (Englewood Cliffs: Prentice-Hall, 1965), 125.
4. Seton-Watson, op. cit., 389.
427
REGIONAL IDENTITY
Like the rest of Eastern Europe, the Balkans has undergone drastic political, social and economic changes since World War II. In addition to the separate cultural factors of national identity, Communist ideology and Russification, one must add the additional process of 'modernization', which is the impact of modern technology as differentiated from a specific ideology. The end result is that each country is undergoing differential development that makes it increasingly difficult to make generalizations for 'the Balkans'. In fact because of the changing regional relations in Europe as a whole it becomes most pertinent to ask "What is the Balkans?"
Theoretically, the Balkan peninsula would be the land south of a line drawn between the northern end of the Adriatic and Black Seas, yet this seemingly ready line of identification does not conform to the political units nor to the ethnic grouping that exist now or in the past. Therefore the precise areal definition of what constitutes 'the Balkans' has been a subjective one based upon different points of view in relation to existing geopolitical situations. The problem is similar to that of the concept of 'Central Europe' (Mitteleuropa or Zwischeneuropa) [5] which has undergone considerable change over a period of time.
Some authors continue to use 'the Balkans' as an area name, but others prefer the term 'Southeast Europe'. [6] This manner of using directional titles has created geographical confusion to the reader because the list of such titles now includes the entire gamut from western, eastern and southern Europe to central, southeast and even east-central Europe. [7] The ultimate value of these multiferous sub-divisions is debatable, but what they do indicate is that Europe is a complex entity which can be studied from different points of view and with different geographic orientations. It is even possible to take an ideological point of view and use an area designation of 'the people's democracies' of Europe focusing on the Communist-dominated countries west of the Soviet Union. [8]
5. George W. Hoffman, A Geography of Europe, 3rd ed. (New York: Ronald Press, 1969), 272-276.
6. R. R. Betts, Central and South East Europe, 1945-1948 (London and New York: Royal Institute of International Affairs, 1950), 227 pp.
7. R. H. Osborne, East-Central Europe: An Introductory Geography (New York and Washington, D.C.: Praeger), 384 pp.
8. Ludwik Straszewicz, ed., Geografia gospodarcza europejskich krajów demokracji ludowej (Warszawa: Pánstwowe Wydawnictwo Ekonomiczne) : Part I, German Democratic Republic, Czechoslovakia. Hungary (1966), 355 pp.; Part II, Rumania, Bulgaria, Yugoslavia, Albania (1968), 441 pp.
428
Not only is there a question as to what name to use for the region, there is also discrepancy as to what specific countries to include in the Balkans. The narrowest definition would be to limit the Balkans to the three countries of Yugoslavia, Bulgaria and Albania. [9] Other authors include Romania [10] and Greece [11] as integral parts of the region. Some justification may also be made that Hungary is in many ways a Balkan culture and that, since European Turkey is part of the Balkan peninsula and since the history of the Balkans is so closely linked to the Turkish empire, perhaps all of Turkey should also be added. [12] Thus the gamut of choice ranges from the smallest nucleus of Yugoslavia, Bulgaria and Albania, to the greater Balkans concept of Hungary, Romania, Yugoslavia, Bulgaria, Albania, Greece and Turkey.
Without attempting to make a definitive area definition, it should be noted that the use of the entire area as a general unit best shows the trends and scale of significant political, economic and social changes that have characterized the Balkans since World War II.
SIGNIFICANT DEMOGRAPHIC TRENDS
The most significant demographic trend in the Balkans since World War II has been the decrease in the birth rate with a corresponding decrease in the annual increase in population. Prior to World War II, western Europe had lower rates of birth than did eastern Europe, but in the postwar period, the situation became reversed with the highest growth rates in western Europe and smaller rates in the rest of Europe.
In the 1960-1967 period, the average rate of population increase for Europe was 0.9 %, but that of western Europe was 1.1%, while southern Europe was 0.9%, eastern Europe, 0.6% and northern Europe, 0.7 %. [13] In this United Nations classification, some of the Balkan countries are
9. George W. Hoffman, The Balkans in Transition (Princeton: Van Nostrand, 1963), 124 pp.
10. Dragos D. Kostich, The Land and People of the Balkans: Albania, Bulgaria, Rumania, Yugoslavia (Philadelphia: Lippincott, 1962), 160 pp. Also Wolff, op. cit.
11. Stavrianos, op. cit. Also Charles and Barbara Jelavich, The Balkans.
12. Charles and Barbara Jelavich, eds., The Balkans in Transition: Essays on the Development of Balkan Life and Politics since the Eighteenth Century (Berkeley and Los Angeles: University of California Press, 1963), 451 pp.
13. United Nations, Statistical Office: Demographic Yearbook, 1967 (New York: United Nations, 1968), 97.
429
grouped in the eastern Europe category, namely Bulgaria, Hungary and Romania, while Albania, Greece and Yugoslavia are placed in southern Europe and Turkey is listed under southwest Asia.
Although the regional trend in the Balkans is toward a decrease in annual population increase, there is still considerable difference in the individual country rates. In the 1963-1967 period, Hungary showed only a 0.3 % annual increase, one of the smallest in Europe. [14] The extent of the decline in Hungary is highlighted by the fact that in the 1953-1959 period, it was 0.6%. [15] Three countries showed an intermediate grouping: Romania at 0.6%, and Bulgaria and Greece both at 0.7%, still below the average for Europe. But Yugoslavia had a rate of 1.2% and Turkey and Albania had the very high rates of 2.5 and 2.8%, respectively. [16]
These rates are directly associated with the corresponding natural increases based on the relationship of rates of birth and death. In 1967, Hungary had a natural increase of 3.8 per thousand resulting from a crude birth rate of 14.5 and a high death rate, 10.7 per thousand. Bulgaria was next with 6.0, based on crude birth of 15.0 and crude death rate of 9.0 per thousand. Greece followed with a birth rate of 18.5 and death rate of 8.3 to create a natural increase of 10.2 per thousand. Yugoslavia showed a somewhat higher natural increase of 10.8 per thousand with a birth rate of 19.5 and death rate of 8.7 per thousand. But Romania had a higher birth rate of 27.1 and a death rate of 9.3 to produce a natural increase of 17.8 per thousand. In contrast to all of these, in 1966 Albania had a birth rate of 34.0 per thousand and a low death rate of 8.6 to make a natural increase of 25.4 per thousand. And Turkey had the even higher birth rate of 43 per thousand with a death rate of 16.0, thus having the highest natural rate of increase of 27 per thousand. [17]
The extent of the decline in both birth and death rates is best evidenced by comparative statistics in previous periods. In 1881-1885 Hungary had a birth rate of 45.0 per thousand and a death rate of 33.5. By the 1921-1925 period, this had dropped to 29.4 and 19.9. In sharp contrast, this fell in 1962 to a birth rate of 12.9 per thousand and a death rate of 10.8 to create the phenomenally low natural increase of 2.1 per thousand. An even more spectacular decline is shown in the deaths of infants less
14. Ibid., 111.
15. United Nations, Statistical Office: Demographic Yearbook, 1960, 113.
16. United Nations: Demographic Yearbook, 1967, 111.
17. Ibid., 121 and 122.
430
than a year old. In the 1901-1905 period the incidence was 217.8 per thousand; in 1967, only 37 per thousand. [18]
In 1890, Bulgaria had a birth rate of 34.9 per thousand and a death rate of 21.0. By 1920, this had increased to 39.9 and 21.4. By 1966, the birth rate had dropped to a peace-time low of 14.9 with a death rate of 8.3 per thousand. [19]
Yugoslavia also reflects the decline in both rates with a rather even decline since 1921 from a birth rate of 36.7 per thousand and a death rate of 20.9 to 18.9 and 8.6 respectively in 1968. [20]
Romania showed the same pattern of constant decline in both birth and death rates until 1966. For example, in 1930 the birth rate was 34.1 per thousand and the death rate, 19.3, which dropped in 1966 to 14.3 and 8.2. But in 1967 and 1968 the birth rate nearly doubled and the death rate also rose to 9.3 and 9.6. [21] Curiously enough, the same change is reflected in the deaths of infants less than a year old. In 1930, this was 175.6 per thousand which by 1965 had dropped to 44.1. But in 1966 and 1967 this rose to 46.6 and jumped in 1968 to 59.5 per thousand. [22]
Greece shows a somewhat different pattern. In the immediate period after World War I, the birth rate was about 21.0 per thousand, then by 1936 moved up to 30.0 and stayed high until the end of World War II, when a decline set in, to a low of 17.7 per thousand in 1965. [23] A great difference also appears in the death rate, in that before World War II, the rate ranged between 13.0 and 18.0 per thousand, but by the end of World War II had dropped to nearly 8.0 per thousand and has stayed near the same level since then. [24] In the years 1963 and 1964, the combination of low birth rate, low death rate and an increase in emigration produced the very low population increases of only 0.37% and 0.36%, among the very lowest in Europe. [25]
Albania shows even greater contrast to these previous situations in
18. Központi Statisztikai Hivatal: Statisztikai Évkönyv, 1967 (Budapest: Statisztikai Kiadó Vállalat, 1968), 10.
19. Narodna Republika B'lgariya, Tsentralno Statistichesko Upravlenie Pri Ministerskiya S'vet: Statisticheski Godishnik na Narodna Republika B'lgariya, 1968 (Sofiya), 21.
20. Socijalistićka Federativna Republika Jugoslavia, Savezni Zavod za Stastistiku: Statistički Godišnjak Jugoslavije, 1969 (Beograd, 1969), 82.
21. Romania, Direcţia Centrală de Statistica: Anuarul Statistic al Republicii Socialiste România, 1969 (Bucuresti, 1969), 69.
22. Loc.cit.
23. National Statistical Service of Greece: Statistical Yearbook of Greece, 1968 (Athens, 1969), 27.
24. Loc.cit.
25. Greece: Statistical Yearbook of Greece, 1968, 17.
431
that the birth rate since World War II is higher than it was before World War II. In 1938 it was 34.7 per thousand and dropped to 28.1 in 1941. In 1946 it was at a low of 27.1 and rose to 43.8 in 1955, then slowly fell to 35.3, still a very high figure for Europe. During the same period, the death rate decreased from 17.8 in 1938 to a low of 8.4 per thousand in 1967. Thus Albania has a high birth rate and a rather low death rate to produce a high natural increase, which in 1967 was 26.9 per thousand. [26] These raw statistics raise questions that are both intriguing and difficult to answer. In the case of Hungary, the high death rate in the past decade is clearly linked to the large exodus of young people during the 1956 revolution thereby leaving a larger proportion of older people in the population structure. But it is more difficult to explain why Albania and Greece have death rates that are only half as high as that of Turkey.
It is also clear that, for the area as a whole, population losses during World War II and the smaller birth rate during World War II are significant factors in the general post-war decline. Similarly the recent availability of contraceptive devices and the possibility of legal abortion also help explain the general decline in birth rates but do not provide clear answers as to why the differences between individual countries. Nor can one attribute these differences primarily to political ideology, because Hungary, a Communist country, and Greece, a non-Communist country, both have low birth rates compared to Albania, a Communist country, and to Turkey, a non-Communist country.
Perhaps what is needed is not to explain each deviation or change year by year but rather to focus upon long range trends and to place them within a conceptual framework that relates to the end result, rather than to focus on individual population factors. Clearly, there are involuntary controls of population such as war or any number of diseases that an individual cannot himself choose to avoid. But just as clearly there are voluntary controls of population that an individual may choose or not choose, such as contraceptive devices and abortions. This voluntary element of deciding whether to have children or not to have children might best be designated as the 'family climate'.
In the past it appears that circumstances in the Balkans were conducive to the creation and maintenance of large families. Contraceptives were difficult to obtain, if not completely non-existent. The high death rate of infants and of still-births encouraged the birth of more children. Farm life itself encouraged large families not only for labor but for companion-
26. Republika Popullore E Shqipërise, Drejtoria E Statistikës: Vjetari Stalistikor i R. P. Sh., 1967-1969 (Tiranë, 1968), 32.
432
able family life. Large-scale emigration, especially at the turn of the century, was another incentive for large families on the assumption that even if some left, there would still be children in the 'old country'. Furthermore church and state encouraged family-hood as did cultural traditions, because in much of the Balkans, perhaps in all the Balkans, manhood was associated with family virility — a man who had fathered many children deserved village respect for that if for no other reason. Thus for centuries the 'family climate' favored large families.
Yet it must be also recognized that some of these same factors in certain circumstances favored less childbirth. In Greece, for example, the long history of emigration has been a negative factor because parents recognized the probability that their children would emigrate permanently overseas.
But over the past half century it appears that regional environmental factors have tended to create a 'family climate' favoring small rather than large families. Some of the countries, like Hungary for example, have not only made it easy to obtain contraceptives but have also permitted legal abortion. There also seems to be some correlation between rise in standard of living and drop in family size. In Yugoslavia in 1968 the lowest birthrates were in the Voivodina, 13.8 per thousand; Serbia Proper, 14.9; Croatia, 15.2; and Slovenia, 17.4; compared to 23.1 in Bosnia-Hercegovina; 25.6 in Macedonia and 37.4 in the Kosovo. Conversely there are more divorces in the areas of higher standard of living ranging, in 1968, from 179.4 divorces per thousand marriages in the Voivodina to only 26.3 per thousand marriages in the Kosovo. If the marriage pattern conformed to this premise, one could predicate a classical 'model' conducive to small families, but actually the marriage rate is the opposite of what one might expect — the highest being 9.0 per thousand inhabitants in both the Voivodina and in Serbia Proper in comparison to only 7.3 in the Kosovo and 6.7 in Crna Gora. [27] Part of the explanation for this lies in the fact that many of the divorced people remarry thus raising the marriage incidence.
Another factor that must be considered in population increase is the age of marriage of the bride and groom. In Bulgaria in the past decade the largest category of brides was under the age of 19, with almost an equal number being married in the 20-24 year old category. The leading category for grooms was the 20-24 year old bracket with the 25-29 year old unit next with about half the number of marriages. [28] Rumania
27. Jugoslavia: Statistički Godišnjak Jugoslavie, 1969, 332-333.
28. Bulgaria: Statistical Yearbook 1968, 11.
433
shows the same pattern of brides mainly below the age of 20 and grooms in the 20-24 category. [29] The same is true of Yugoslavia with the brides generally marrying at a younger age than the men. [30]
But in Greece both men and women marry at a later age, similar to the situation in Ireland. In 1967, the largest category for women was the 20-24 year group totaling some 31,595 out of 81,706 marriages. The next largest category was the 25-29 group, which was just slightly larger than the 15-19 group. For men, the predominant group was 25-29 years of age totaling 26,705 out of the 81,706 marriages. The next group was 30-34, which was slightly larger than the 20-24 group. [31] Thus, in Greece, clearly both men and women marry some five to ten years later than in the other Balkan countries.
Why then do Greek men and women marry at a later age? The question is easier to ask than to answer, but some causes seem reasonably clear. The poverty of the countryside is often quoted by peasants as a reason to delay marriage and also the fact that the young men in great part either emigrate out of Greece or go to the towns so that the villages suffer a dearth of eligible young men. Another reason that is given is that a woman is expected to bring a sizeable dowry to the marriage. Not only does it take time for a woman to accumulate a large dowry, but an older man who is well established can demand more than a younger man, hence the dowry system tends to perpetuate older marriages.
Statistically, marriages in Greece were indicated in 1967 in three categories : urban, semi-urban and rural. Out of a total of 80,765 in all Greece, 44,056 were urban, 8,841 semi-urban, and 27,868 rural. This would indicate an urban predominance, but this is a very inconclusive generalization because more than half of the urban marriages, 26,522, were in Athens. And actually outside of Athens itself, rural marriages predominated over urban marriages even in Macedonia, where Salonika with its large population had a large number of marriages. [32] Since the age of marriage in the cities is also comparatively older as it is in the countryside it is probable that the low natural rate of increase in Greece will continue as it is.
The relationships of ethnic groups to population increase must also be raised. This question is particularly pertinent to the Balkans where each country has large numbers of minority groups. The large number
29. România: Anuarul Statistic al Republicii Socialiste România, 1967, 98-99.
30. Jugoslavia: Statistički Godišnjak Jugoslavie, 1969, 340.
31. Greece: Statistical Yearbook, 1968, 29.
32. Ibid., 28.
434
of ethnic groups in the Balkans precludes detailed discussion here but because the administrative divisions of Yugoslavia, the republics, are based on ethnic grounds, they well illustrate ethnic differences, especially in terms of natural increase. In the 1950-1954 period the natural increase per thousand inhabitants by republics was as follows : Bosna-Hercegovina, 24.3; Crna Gora, 22.1; Croatia, 11.5; Macedonia, 23.9; Slovenia, 11.9; and Serbia, 15.0, which was further divided into Serbia Proper, 14.8, the Voivodina, 10.9, and the Kosovo, 25.5. In 1968 they were: Bosna-Hercegonina, 16.5; Crna Gora, 14.3; Croatia, 5.1; Macedonia, 17.6; Slovenia, 6.7; and Serbia, 9.3, which was divided into Serbia Proper, 6.6, the Voivodina, 4.6, and the Kosovo, 28.5. [33] This shows that in all administrative units except one there had been sizeable decreases in natural increase but that in that one, the Kosovo populated mainly by Albanians, the natural increase had not only not declined, it had been augmented so that it was some seven times as great as the Voivodina and over five times as great as Croatia.
This example highlights the problem of differential growth of ethnic groups within Balkan countries. In general, the minority groups are growing faster than the majority groups of individual countries. In the future this may increase internal frictions in the countries that have large and well identified minority groups, such as the Albanians in Yugoslavia and the Hungarian and German minorities in Romania.
These demographic trends of decreasing birth and death rates, higher divorce rates and differential ethnic growth rates in the Balkans are all similar to general demographic trends in the world at large. Yet the basic fact remains that in the Balkans as in the world there is still an increase of population each year, and, even at these reduced rates, there is growing population pressure in all the Balkan countries. This is being expressed both by increasing area densities and by internal migrations, mainly to urban areas.
URBAN MIGRATION
All of the Balkan countries show increasing density of population. Bulgaria had a density of 48 persons per square kilometer in 1921; 60.9 in 1941 and 75.2 in 1967. [34] Yugoslavia had 49.0 in 1921, 56.8 in
33. Jugoslavija: Statistički Godišnjak Jugoslavije, 1969, 333.
34. Bulgaria: Statisticheski Godishnik na Narodna Republika B'lgariya, 1968, 19.
435
1948 and 72.5 in 1961. [35] In 1821 Greece had only 19.8 per square kilometer, 39.5 in 1920, 56.8 in 1940 and 63.6 in 1961. [36] Rumania had 60.1 in 1930, 66.8 in 1948 and 80.4 in 1966. [37] Hungary has had the highest with 53.9 in 1869, 85.9 in 1920, 100.1 in 1941 and 107.1 in 1960. [38] Albania showed the greatest post-war increase in density from 39 in 1945 to 68 in 1967. [39] Turkey has the lowest country density of 42 persons per square kilometer but if a geographical division is made of Turkey, a surprising contrast appears because Asiatic Turkey has only 40 per square kilometer but the European portion has the very high density of 118 persons per kilometer. [40]
Although these country densities offer some comparisons in terms of population growth, the reality is that there is geographical concentration of people into a relatively small portion of each country. This is highlighted by Greece where it is estimated that more than a third of the population live on less than 5 % of the land and about a half live on less than 15% of the total territory. [41]
This increasing concentration of population is best indicated by the changes in the ratio of urban to rural population. But these ratios must also be used with caution because each country makes its own definition of what constitutes urban and rural classification. Not only do these definitions often change with different censuses, but some of the countries have added a third classification, semi-urban, which is hard to combine with either urban or rural.
As might be expected, Albania is still predominantly rural. In 1923 the ratio was 84.1 % rural and 15.9% urban. By 1967, this had become 66.7% rural and 33.3 % urban. [42] Bulgaria in 1920 had an urban ratio of 19.9 % [43] but by 1965 nearly half the country was classed as urban, 46.4 %. [44]
In the census of 1961, Greece introduced a three-category classification, wherein 'urban' was defined as municipalities and communes in which the largest center had a population of 10,000 or more inhabitants. The
35. Jugoslavija: Statistički Godišnjak Jugoslavije, 1969, 82.
36. Greece: Statistical Yearbook, 1968, 16.
37. Romania: Anuarul Statistic al Republicii Socialiste România, 1967, 67.
38. Hungary: Statistical Yearbook, 1965, 7.
39. Albania: Vjetari Statistikor i R. P. Sh., 1967-1968, 3.
40. United Nations: Demographic Yearbook, 1967, 109.
41. Bernard Kayser, Géographie Humaine de la Grèce (Paris: Presses Universitaires de France, 1964), 15.
42. Albania: Vjetari Statistikor i R. P. Sh., 1967-1968, 27.
43. United Nations: Demographic Yearbook, 1960, 385.
44. Bulgaria: Statistical Yearbook, 1968, 3.
436
'semi-urban' category was for those municipalities with a largest population center between 2,000 and 9,999 inhabitants, and the 'rural' category was for those in which the largest center had less than 2,000 inhabitants. Because semi-urban is a statistical rather than functional division, it makes it difficult to include this in either the strictly urban or strictly rural category. Using the 1961 classification, in 1928 the urban percentage was 31.1; semi-urban, 14.5%; and rural, 54.4%. Corresponding figures for 1961 were 43.3% urban; 12.9%, semi-urban; and 43.8%, rural. [45] This shows the increase in urban population and the decrease of both the semi-urban and rural categories.
For Romania the proportion in 1930 was 21.4% urban and 78.6% rural population. By 1956 this had changed to 31.3% urban and 68.7% rural and in 1966 to 38.2% urban and 61.8% rural. [46]
But for Yugoslavia the urban ratio is still surprisingly small. According to the censuses of 1948 and 1953, it was only 16.2 and 18.5 %. [47] By the 1961 census, this had increased to only 28.0 %. [48] Even Turkey had a higher ratio of urbanization, which rose from 24.2% in 1927 [49] to 32.2% in 1960 and 34.4% in 1965. [50]
Again these raw statistics must be viewed with caution. Actually the process of urbanization is not quite the same as one imagines in western Europe and the United States. Instead, in the Balkans there is a marked functional difference between the city as represented by a small agricultural segment of population and the large rural agglomeration which is primarily agricultural in function but statistically is large enough to surpass the usual urban definition of any settlement of more than two thousand inhabitants.
Therefore some Balkan 'cities' are, in reality, large agricultural villages. It is clear that all the Balkan countries are undergoing complex internal migrations, of which one product is urbanization. But, most unfortunately, there is a dearth of factual data available on the specific details of individual migration. While it is possible statistically to see the end results of migration, it is difficult, if not actually impossible, to see the processes of movement involved. There is a multiplicity of cross-migrations not only from village to village and village to city, but also
45. Greece: Statistical Yearbook of Greece, 1968, 21.
46. Romania: Anuarul Statistic al Republicii Socialiste România, 1967, 67.
47. United Nations: Demographic Yearbook, 1960, 389.
48. United Nations: Demographic Yearbook, 1967, 198.
49. United Nations: Demographic Yearbook, 1960, 385.
50. Turkey, State Institute of Statistics: Statistical Yearbook of Turkey, 1969 (Ankara: State Institute of Statistics Printing Division, 1969), table 1.
437
from city to city and from city back to the village. Distinctions must also be made as to area migrations both within regions and between regions.
The extent of such migration is evidenced by studies of the 1961 census in Yugoslavia which indicate that only 7% of the 340 towns had at least two-thirds of the inhabitants born in the town, that some 60% of the towns had between one-third and two-thirds of the population native-born, and that 33 % of the towns had a native-born population of less than one-third. Stated conversely, nearly one-third of all the towns in Yugoslavia had two-thirds of their inhabitants born elsewhere. [51]
Not only are towns growing by migration from the countryside, there is some indication that birth rates in the towns have not decreased as fast as in the countryside, so that in fact there is somewhat of a greater natural increase in the cities, thus accentuating the growth of cities. In Yugoslavia, for example, there was a 14.0 per thousand inhabitants increase in total population between the 1948 and 1953 censuses, but the urban increase was 38.2 per thousand. There was an even greater increase from the 1953 census to the 1961 census where the total population increase was 10.9 per thousand but the urban increase was 43.5 per thousand. [52]
This trend is also illustrated by Bulgaria. Until 1948 the natural increase in the countryside was higher than in the cities. In 1939 the rural increase was 9.0 per thousand population compared to the urban rate of 4.7 per thousand. By 1948 this difference had narrowed to 12.2 for the rural areas and 11.6 for cities. In 1949 this reversed to 13.2 urban and 12.8 rural and the difference has grown progressively to 8.9 urban and 3.3 rural in 1967. [53]
This trend can also be expressed by the specific growth of individual cities in Bulgaria. In the period 1946 to 1965, the population increase in Bulgaria was 15.2 %. But the population of the city of Sofia increased by 69 %, nine cities increased over 200 % and 36 cities increased between 100 and 200%. By contrast, only three cities decreased in population. [54]
Urbanization is, then, one of the striking demographic trends in the Balkans. But the different countries show quite different areal patterns of urbanization.
51. Ivanka Ginić, Dinamika i Struktura Gradskog Stanovništva Jugoslavije (Beograd: Štampa RadiSa Timotić, 1967), 25.
52. Ibid., 130.
53. Bulgaria: Statistical Yearbook, 1968, 10.
54. Ljubomir Dinev, "Nyakoi Izmeneniya v Razpredelenieto na Naselenieto na N. R. Bʹlgariya", Geografiya, Godina XVI (1966), Knizhka 1, 4.
438
PATTERNS OF URBANIZATION
The towns of the Balkans are not new. Most of them date back not only to medieval times but to the classical period as well. Thus history has helped mold urban characteristics over a long period of time. Secondly, ethnic groups have occupied their present territories over a considerable span so that ethnic influences are related to urban structure. Thirdly, most of the Balkan countries have existed for some time, and even though Yugoslavia and Albania date back only to the Balkan Wars of 1912-1913, certain territorial units had administrative continuity, such as Bosnia-Herzegovina over centuries, so that there has been sufficient time for a political imprint to be implanted on the urban scene. Therefore the general areal pattern of towns today represent a historical selection and development not only in terms of political, social, economic and military factors but of the attractions or limitations of natural conditions as well. Therefore although it must be recognized that each country has its own mosaic of areal distribution of cities, three distinct patterns are recognizeable.
Greece constitutes one type in which a prime city dominates the country. According to the census of 1961, Athens alone had 52% of the urban population of Greece and 22% of the total population of the whole country. [55] The second largest city is Salonika, which in 1961 had only some 20% of the population of Athens. In addition to these two, there is only one other city with a population of over 100,000 people. That is Patras, whose conurbation included some 102,000 compared to the 1,853,709 of Athens and the 378,444 of Greater Salonika. [56] Athens is a classical example of the powers of attraction of a capital city, where all factors, political, economic, social, military, historical, physical, and even its location in the geographical center of the state of Greece, favor constant aggrandizement of the populace. It is difficult to see how regional planning to aid the other cities can hope to cope with the magical allure of the very name of Athens.
Sofia in Bulgaria is also a dominant capital city, but, in 1965 its population of 801,000 represented only 21 % of the urban population and 9.7% of the total population of the country. There were four other Bulgarian cities that had populations of more than 100,000 inhabitants — Plovdiv, Varna, Ruse and Burgas. But there were 27 cities in the category of 25,000 to 99,999 inhabitants and 34 in the 10,000 to 24,999 class, which
55. Bernard Kayser, op. cit., 28-29.
56. Greece: Statistical Yearbook of Greece, 1968, 22.
439
shows the dispersed pattern of cities outside of Sofia. [57] This is the second urban type where there is a prime city but also a number of secondary centers which show a regional influence. Albania with the city of Tirana, Turkey with Istanbul, Romania with Bucharest, and Hungary with Budapest all represent this second urban type.
But in Yugoslavia there is no single dominant city. To be sure Belgrade, the capital, is the largest city, but its population of 585,234 in 1961 formed less than 4 % of the total population. And the second city, Zagreb, was only about 25% smaller with its population of 430,802. In addition there were five other cities with a population larger than 100,000 inhabitants, namely Ljubljana, Novi Sad, Rijeka, Sarajevo and Skopje. [58] This lesser urban concentration may be expressed in another fashion by indicating that of the 348 towns listed in the 1961 census, only 14 had more than 50,000 inhabitants and they included only 41.4 % of the urban population. This means that the majority of Yugoslav towns are relatively small, actually more than two thirds of them have less than 10,000 inhabitants. [59]
In a sense the urban pattern of Yugoslavia is a decentralized one in that many geographical regions contain large local centers. The capital cities of the individual republics highlight this decentralization. They represent location and growth under different administrative situations, but their strength has been their usefulness as local urban centers for different ethnic groups. The creation of the administrative system of republics in postwar Yugoslavia has given them renewed growth and vigor as ethnic capitals.
This concentration of population in the republic capitals is shown by the fact that the eight administrative capitals of Yugoslavia in 1961 contained only 8.8% of the total population but included 31.3 of the total urban population. The respective figures for the republic capitals are: Sarajevo in Bosna-Hercegovina, 4.4 and 22.4; Titograd in Crna Gora, 6.2 and 28.6; Zagreb in Croatia, 10.4 and 33.6; Skopje in Macedonia, 11.8 and 33.8; Ljubljana in Slovenia, 8.4 and 30.8; Belgrade in Serbia Proper, 12.4 and 43.3; Pristina in the Kosovo-Metohija, 4.0 and 20.2; and Novi Sad in the Voivodina, 5.5 and 14.2 %. [60]
POPULATION HEARTLANDS
These three urban types stress the areal patterns of urban development and concentration. But all of the countries also have rural areas that
57. Bulgaria: Statistical Yearbook, 1968, 6-7.
58. United Nations: Demographic Yearbook, 1960, 228.
59. Ivanka Ginić, op. cit., 22.
60. Ibid., 116.
440
show differential densities of both population and settlement. These rural areas may have extremely high density of population and sometimes are located near and associated with major cities. This is the situation near Zagreb, where the nearby Zagorje has the highest density of rural population in Yugoslavia.
In some of the Balkan countries certain areas show such a decided concentration of population, both rural and urban, that they constitute population 'heartlands' on the basis of their demographic, political, economic and social significance.
Greece in particular shows such territorial demographic concentration. Although there is a surprisingly even distribution of peoples and settlements throughout peninsular and insular Greece, on the bases of density of population and on a higher rate of population increase, a zone, sometimes called the 'fertile crescent', extends from the Attic plain of Athens to the plains of Macedonia with the city of Salonika, and projects northeastward into western Thrace. [61] This zone poses the whole question of planned development because the inexorable growth of Athens raises the issue as to whether the government should foster economic development in the poorer regions. The likelihood is that even if the growth of Athens is limited in some form, it is still this 'heartland' zone that will show the greatest growth.
Yugoslavia also has a territorial 'heartland'. This extends in a crescent-shaped arc from the northern Adriatic coast through eastern Slovenia into the Drava and Sava lowlands of the Voivodina and southward through western Serbia into the Kosovo. [62] This is a long area which does not have a consistently high density throughout its span, but this qualitative designation does show the area of increasing population concentration in the past half century. Therefore, it too is the probable zone of greatest future growth.
In Romania, the basic pattern of population is most directly associated with the Transylvanian Mountains, which constitute a zone of low density of population. In contrast the Transylvanian Basin, the Hungarian lowland to the west, the Moldavian plains to the east and the Wallachian plains to the south all contain large urban and rural popula-
61. Bernard Kayser and Kenneth Thompson, Economic and Social Atlas of Greece (Athens, 1964), sections 201-206.
62. Institut Društvenih Nauka, Centar za Demografska Istraživanja, Šema Stalnih Rejona za Demografska Istraživanja (Beograd, 1963), 73-76. Also Dragan Rodić, Geografija Jugoslavije, II Deo. (Beograd: Naučna Knjiga, 1967), 51-52.
441
tions. But the greatest zone of concentration is in the central Wallachian plain west of the cities of Bucharest and Ploeşti. [63]
Hungary which consists mainly of plains areas forms a great contrast to the other countries because of the high density of population throughout almost all of the regions of the country. But Budapest does act as the nucleus of a major area of concentration along the upper Danube. In addition there is a secondary center in the lower Tisza river valley northeast of Szeged. [64]
In Bulgaria the situation is quite the opposite from Hungary because the mountains divide the country into a series of isolated regional centers. The two largest of these are the Sofia basin and the upper Maritsa valley with the city of Plovdiv. [65] Although these are separated by mountains, they do form a population axis which will probably continue to grow at a faster rate than the other parts of the country.
In Albania the settlement pattern is much simpler than in the other Balkan countries. The coastal plain on the west stretching from the town of Shkodër to Vlorë is heavily populated with the greatest density between the coastal port of Durrës and the capital, Tiranë. [66]
CONCLUSIONS
In the period since the end of World War II, new demographic trends are changing markedly the structure and growth of population in the Balkans. New political regimes, changing technology and decreasing isolation all play roles in higher standards of living and generally decreasing birth and death rates. Although there are differential rates of population growth in the different countries, all are experiencing growing urbanization and an increasing concentration of people into population 'heartlands', which are experiencing higher rates of growth than other national regions. These Balkan 'heartlands' appear to offer the most attractive conditions for increased settlement, therefore they constitute the pattern of the future wherein there will be increased concentration in these same regions rather than the creation of new 'heartlands'.
63. Victor Tufescu, ed., Atlas Geografic Republica Socialistă România (Bucuresti: Editura Didactica Si Pedagogica, 1965), 58-61.
64. Márton Pécsi and Béla Sárfalvi, The Geography of Hungary (Budapest: Kossuth, 1964), 163-166.
65. Geografski Atlas na Bʹlgariya (Sofiya: Glavno Upravlenie po Geodeziya i Kartografiya, 1966), 12.
66. Albania: Vjetari Statistikor i R. P. Sh., 1967-1968, 27.